Reflecting on literacy and numeracy progress measures for Māori, Pasifika, New Zealand Pākehā and Other ethnicities for the period 2012-2014 at Waikato Institute of Technology (Wintec)

Greyling, Willfred (2015) Reflecting on literacy and numeracy progress measures for Māori, Pasifika, New Zealand Pākehā and Other ethnicities for the period 2012-2014 at Waikato Institute of Technology (Wintec). Wintec/TEC report, 31 September 2015. (Unpublished)

[img] PDF
Restricted to Repository staff only

1497Kb

Abstract or Summary

In this report, we outline the data-processing we undertook to replicate the TEC’s (2012) LNAT algorithm for calculating statistically significant reading and numeracy gain for students in the institute’s Student Data Return (SDR) files for the period 2012-2014. Using the Pre/Post Score category coding, we selected these students from the full data set referred to in Sub-report 1 (Greyling, 2015a) on patterns of LNAT use at the institute for the period 2011-2014. Our main aim, however, was to use a repeated measures analysis which required a time-intensive process of coding to convert the data sets 1 and 2 (See Sub-report 1) (Greyling, 2015a) into a multivariate layout (Kinnear & Gray, 2011: 22-23; Tredoux, 2002: 317). Using Reder’s (2012) view that significant literacy development occurs over a period of approximately 5 to 6 years, we then argued that time lapse could be used as a variable in the calculations. We reasoned that somehow the time in between initial and progress assessments could be used to differentiate between learners with short (say, semester-long) and longer periods of exposure (say, 18 months under the sequence concept) to literacy development (which would also be affected by natural growth and a host of other variables). We argued that we could work with a proportion of a base period, and to this end, we selected two periods (scenario 1 = 4 years; and scenario 2 = 5 years) that fell within Reder’s (2012) claim of 5-6 years as a realistic period to record LN development; and another which was the optimum period suggested by Reder (2012), plus 1 year (scenario 3 = 7 years). These were the denominators in the fraction, with numerators representing the time lapse between assessments. Then, for each scenario, we modified the TEC (2012) algorithm to include the term (expressed as a proportion of gain score required to achieve statistically significant gain). Our results showed that for a 5-year base period, approximately 70% of students could be deemed to have achieved statistically significant gain. This approach, we concede, is contentious, and was labelled a contestable form of statistical extrapolation by the second reviewer of the research. We report the details, and remain convinced that somehow time lapse should be accommodated in LNAT progress calculations. Alternatively, we recommend that a Gain Index for different course durations and TEC funds be developed, based on 5 years of longitudinal data available in the LNAT.

Item Type:Report
Keywords that describe the item:literacy, numeracy, New Zealand, Maori, Pasifika, Wintec, Waikato Institute of Technology
Subjects:L Education > LC Special aspects of education
Divisions:Schools > Centre for Foundation Studies
ID Code:3911
Deposited By:
Deposited On:12 Oct 2015 03:42
Last Modified:10 Apr 2017 02:33

Repository Staff Only: item control page